Parkinson's disease (PD) is a chronic progressive disabling condition of the brain that affects balance, movement, and non-motor functioning; there are over 40,000 Veterans with PD across VHA. While many evidence-based treatments are available for managing PD symptoms, numerous gaps in care exist. In our prior research, we applied the VA Quality Enhancement Research Initiative framework and (1) developed evidence-based quality indicators for PD care, employing a national expert panel and modified Delphi method; (2) identified gaps in VA and community care relative to these indicators; (3) analyzed variations in PD care delivery; (4) convened a multi-disciplinary VA and community-wide task force that selected 38 PD quality indicators in ten domains and developed patient-centered, site-specific PD care tools and protocols; and (5) created a Chronic Care Model-based nurse-led intervention.
The project objective was to test whether a nurse-led, Chronic Care Model-based coordinated care-management program improved the quality of care and health status for Veterans with PD, and to determine the cost and feasibility of intervention implementation.
In a 5-year randomized controlled trial, we enrolled Veterans with PD that were then randomized in a 1:1 ratio to receive up to 24 months of a PD care management program, "Care Coordination for Health Promotion and Activities in Parkinson's Disease" (CHAPS), or continue with usual care. Eligible participants across 5 sites of the Southwest Parkinson's Disease Research, Education and Clinical Center (SW PADRECC): Greater Los Angeles, Las Vegas, Loma Linda, Long Beach and San Diego VISN 22 (VA Desert Pacific Healthcare Network) facilities were identified through administrative data and vetted by physicians. The multi-faceted nurse-led intervention, CHAPS, incorporated PD quality improvement (QI) tools to enhance proactive Veteran-centered care delivery. PDQI tools included: (1) a structured telephone assessment administered by a nurse care manager (NCM) to proactively identify medical problems and unmet needs; (2) problem-specific evidence-based interventions organized into treatment plans including follow-up/monitoring protocols with VA and community linkages; (3) the Siebens Domain Management ModelTM, a structural approach to improve collection and organization of health information and enrich communication among providers; (4) Siebens Health Care Notebooks prepared with Veteran self-management materials; (5) a Microsoft Access care management database, containing the CHAPS Assessment and algorithms for problem identification based on Veteran data; and (6) decision-support meetings of NCMs with movement disorder specialists. Research interviewers administered a structured telephone survey at baseline, 6, 12, 18, and 24 months. Participants were provided $25 per survey. Medical records were abstracted.
Analyses (1) measured PD quality indicator adherence, health outcomes, and health service utilization; (2) evaluated qualitative and quantitative data to summarize the extent of intervention implementation and barriers and facilitators to potential dissemination; and (3) examined costs of CHAPS care and cost offset between randomization arms.
Enrollment commenced August 2012 and ended October 2015. A total of 345 Veterans were enrolled (GLA 84, Las Vegas 20, Loma Linda 87, Long Beach 42, and San Diego 112). A total of 328 subjects completed the baseline survey and were randomly assigned to the intervention (162) or usual care (166). Average age was 71 years old, 78% were Caucasian race, 77% had more than a high-school education, 65% were retired and 23% were not able to work because of a health disability.
The primary outcome was adherence to 38 PD indicators. A total of 18 indicators were measured. Of the 18, 6 were measured using patient survey data; the other 12 were measured through chart review. Secondary outcomes were measured using patient survey data.
A total of 281 patients completed the study (143 in the usual care arm and 138 in the intervention arm). Among the 23 usual care and 24 intervention patients who did not complete the study, the reasons included death (n=9 in each arm), withdrawal from the study by the patient (n=7 in usual care and n=10 in the intervention arm), and physician decision (n=7 in usual care and n=5 in the intervention arm).
The primary outcome, adherence to PD indicators at 18 months, was 58% in the usual care arm and 77% in the intervention arm (mean difference, 19%; 95% confidence interval, 16% to 22%), a statistically significant difference favoring the intervention arm. All-cause mortality did not differ between the two arms. There were no serious or other adverse events.
Among nine key secondary outcomes, there was no statistically significant difference observed except in one outcome: the Patient Health Questionnaire 2 (PHQ-2, a screening measure for depression). For the PHQ-2, the percent of patients with a score of at least 3 stayed roughly stable in the intervention arm but worsened significantly in the usual care arm, resulting in a difference in change score of -13% (95% confidence interval, -25% to -2%), favoring the intervention arm. This result was robust to several sensitivity analyses.
Medical record review showed no significant difference in utilization of face-to-face care between intervention and usual care arms. As expected given the telephonic nature of the CHAPS intervention, telephone calls were much more common in the intervention arm than the usual care arm (mean 3.02 calls in the intervention arm compared to 0.11 in the usual care arm, p<0.05).
CHAPS Program implementation fidelity measures included availability of nurse care managers (NCMs) and neurologists. Intervention fidelity (adherence to CHAPS procedures) was assessed through medical record abstraction of NCM documentation. Intervention design details were published (Connor et al., Dec 2015).
Assessment of sustainability and scale-up/spread potential and data were examined through Veteran and clinician impressions and experience. These included type and frequency of Veteran problems, NCM actions, and Veteran activities, impressions and use of MyHeatheVet, types of non-VA sources of care, and counts of education sheet handouts. Qualitative impressions and experiences of 28 Veterans, 7 NCMs, and 10 neurologists were examined through analyses of focused survey data collected at study termination. The usability of Siebens Domain Management ModelTM and the Siebens Health Care Notebook were assessed through a card sorting method of NCM questionnaire data. Results of these analyses showed a promising future for CHAPS.
This study demonstrated the feasibility and implementation potential of a novel, proactive nurse-led coordinated care-management program to provide Veteran-driven, reliable, high-quality PD care. Results suggest positive impacts on care quality. Qualitative feedback from NCMs, neurologists, and Veterans enrolled in the CHAPS Program was generally positive as well, suggesting that Veterans with PD could benefit from the program.
Ideally the integration of care management software within VA's new electronic medical record system could enhance this program. However, the Program can be disseminated without additional IT software and programming. The CHAPS Program, which was designed to accommodate multimorbidity, did so through care coordination with other specialties and VA social services; staffing and resource implications of CHAPS for these other disciplines must be considered. Veterans were given choices to use community resources and did so frequently; future implementation of the program will require assessment of the availability of such resources for local Veterans.
The CHAPS participating sites have considered ways to implement the CHAPS Program, tailoring to their individual sites. Research and/or quality improvement could further evaluate dissemination efforts and enhance CHAPS effectiveness regarding its benefits, costs, and other resource requirements. Positive findings from this additional work could support a potential future decision in favor of large-scale implementation across VHA.
- Connor KI, Siebens HC, Mittman BS, McNeese-Smith DK, Ganz DA, Barry F, Edwards LK, McGowan MG, Cheng EM, Vickrey BG. Stakeholder perceptions of components of a Parkinson disease care management intervention, care coordination for health promotion and activities in Parkinson's disease (CHAPS). BMC neurology. 2020 Dec 2; 20(1):437.
- Connor KI, Siebens HC, Mittman BS, Ganz DA, Barry F, Ernst EJ, Edwards LK, McGowan MG, McNeese-Smith DK, Cheng EM, Vickrey BG. Quality and extent of implementation of a nurse-led care management intervention: care coordination for health promotion and activities in Parkinson's disease (CHAPS). BMC health services research. 2020 Aug 10; 20(1):732.
- Connor KI, Cheng EM, Barry F, Siebens HC, Lee ML, Ganz DA, Mittman BS, Connor MK, Edwards LK, McGowan MG, Vickrey BG. Randomized trial of care management to improve Parkinson disease care quality. Neurology. 2019 Apr 16; 92(16):e1831-e1842.
- Connor K, Cheng E, Siebens HC, Lee ML, Mittman BS, Ganz DA, Vickrey B. Study protocol of "CHAPS": a randomized controlled trial protocol of Care Coordination for Health Promotion and Activities in Parkinson's Disease to improve the quality of care for individuals with Parkinson's disease. BMC neurology. 2015 Dec 15; 15:258.